A public chatbot may receive prohibited topics and personal information.
Apply policy to both inputs and outputs.
Which operational design is most appropriate?
Combine input/output checks, policy rules, monitoring, minimized logs, and human escalation for exceptional cases.
Detailed explanation
A prompt alone cannot guarantee safe behavior for every input and output.
Layered controls are more robust than relying on one prompt instruction.
Logs also require minimization, access control, and retention design.
Without measurement, controls cannot be verified or improved.
Check official documentation
Compare the following points with the official documentation.
Amazon Bedrock Guardrails公式ドキュメントとAWS Well-Architectedの安全性設計を確認する。
Official references
What to confirm
安全制御をポリシー・検査・監視・人手対応の多層構成で説明できる。
Key points
- Input and output controls
- Monitoring
- Escalation
Notes
- Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Defense in depth
Independent controls reduce the impact of a failure in any one control.
Operational feedback
Logs and safety metrics show whether a control works and where it needs adjustment.